Effect of impact exercise on bone mineral density in elderly women with low BMD: a population-based randomized controlled 30-month intervention.

Evidence of the effect of exercise on bone loss comes mainly from studies in voluntary postmenopausal women, and no population-based, long-term interventions have been performed. The purpose of this population-based, randomized, controlled trial was to determine the effect of long-term impact exercise on bone mass at various skeletal sites in elderly women with low bone mineral density (BMD) at the radius and hip. Participants (n=160) were randomly assigned to 30 months either of supervised and home-based impact exercise training or of no intervention. The primary outcome measures were femoral neck, trochanter and total hip BMD, and the secondary outcomes were bone density measures at the radius and calcaneum. Outcomes were assessed at baseline, 12 months and 30 months using blinded operators. The analyses were performed on an intention-to-treat analysis. Mean femoral neck and trochanter BMD decreased in the control group [-1.1%, 95% confidence interval (CI) -0.1% to -2.1% and -1.6%, 95% CI -0.4% to -2.7%], while no change occurred in the exercise group. Mean trochanter BMC decreased more in the control group (-7.7%, 95% CI -9.7% to -5.6% vs. -2.9%, 95% CI -5.3 to -0.9). There were six falls that resulted in fractures in the exercise group and 16 in the control group during the 30-month intervention (P=0.019). A significant bone loss occurred in both groups at the radius and calcaneum. In multivariate analysis, weight gain was associated with increased BMD and BMC at all femur sites both in the exercise group and in the pooled groups. In conclusion, impact exercise had no effect on BMD, while there was a positive effect on BMC at the trochanter. Exercise may prevent fall-related fractures in elderly women with low bone mass.